master thesis: drilling engineering well control with
TRANSCRIPT
Master thesis: Drilling engineering
Well control with
Controlled Mud
Pressure Reviewing AGR’s well control procedures for the CMP system, analyzing and comparing
with conventional well-control procedures., possible impact of technology.
Morten Hansen
14.06.2011
II
Norges teknisk-naturvitenskapelige universitet Fakultet for ingeniørvitenskap og teknologi Studieprogram i Geofag og Petroleumsteknologi Faculty of Engineering and Technology Study Programme in Earth Sciences and Petroleum Engineering
Institutt for petroleumsteknologi og anvendt geofysikk Department of Petroleum Engineering and Applied Geophysics HOVEDOPPGAVE / DIPLOMA THESIS / MASTER OF SCIENCE THESIS Kandidatens navn/The candidate’s name: Morten Hansen Oppgavens tittel, norsk/Title of Thesis, Norwegian: Brønnkontroll med AGRs dual
gradient system Oppgavens tittel, engelsk/Title of Thesis, English: Well control with Controlled Mud
Pressure Utfyllende tekst/Extended text: Background AGR has proposed and patented and MPD system called Controlled Mud Pressure (CMP) with a cost reduction potential for deep water drilling. The mud return is taken out of the riser and pumped up to surface with a subsea mud pump. The riser is partly evacuated and the downhole pressure can be controlled by adjusting the mud level in the riser. Well control is one of the challenges with this new system, since conventional well control can not be applied. Tasks: 1) Literature study: dual gradient systems in general and CMP in particular. Identify equipment and procedures\software needed. Discuss challenges and benefits. 2) Conventional well control for floaters: Discucc conventional well control for deep water drilling from floaters with subsea BOP. 3) Automatic well control with MPD: Discuss the automatic well control solutions offered by AtBalance and Secure Drilling 4) Well control with CMP: Define and discuss kick\loss detection, shut-in procedures and how to circulate out a kick. Discuss how to handle faults on the equipment during a kick. 5) Discuss well control with CMP compared to conventional well control and backpressure MPD.
Supervisor: John Morten Godhavn Studieretning/Area of specialization: Petroleum Engineering / Drilling
Technology Fagområde/Combination of subject: Boring / Drilling Tidsrom/Time interval: January 17 – June 14, 2011
XJohn-Morten Godhavn
IV
Summary
Dual Gradient Drilling (DGD) are a portfolio of drilling systems that in various ways pump
returns from the well through a mud return line (MRL) from the seafloor, or somewhere
between the seafloor and the rig floor. Two variations exist, with or without a marine riser.
The system without riser is an open system which is intended for tophole sections where a
blowout preventer (BOP) is not used. When depths are reached that require a BOP, a subsea
BOP can be applied together with a riser. Both systems results in the augmentation of the
fluid column by adding a lighter fluid gradient on top of the outtake to the MRL and subsea
pump. In the event of an open system, this will be the seawater, with riser different blanket
fluids can be used, like seawater, nitrogen or just a lighter mud component like base oil.
The controlled mud pressure (CMP) system is offered by AGR and applies a MRL together
with a subsea pump to take returns from the wellhead to the rig. This system uses a riser,
and can thus drill deeper than tophole applications and is intended to handle well control
situations. Together with systems such as @Balance’s dynamic annular pressure control
(DAPC) system and Secure Drilling’s microflux control (MFC) it represents the future of
automated kick detection, control and handling. The two latter are offered as parts or
supplements to mainly managed pressure drilling (MPD) systems. One of them also aims to
increase safety by automation to all drilling systems on any rig.
The CMP technology can be seen as enabling and improving for drilling deepwater
prospects, as well as through zones with either tight drilling windows or depleted pressures.
The potential for automated well control can become a major increase in safety and improve
the way kicks are handled. There are some challenges still to overcome, mainly the
development of the anti u-tube valve, or other options to solve problems related to u-
tubing, as well as looking into the issue of gas boiling in the circulation system.
VI
Preface
This Master’s Thesis has been completed at Norwegian University of Science and Technology
(NTNU) in cooperation with Statoil and also AGR Drilling Services. During this work my
supervisor has been John-Morten Godhavn, as a representative for both NTNU and Statoil.
The work done for this report was outlined by Godhavn, and has been altered along the way
to conclude in the following report. The thought behind the work has been to analyze the
well control of a newly developed and promising drilling method. The aim was to analyze
and simulate the different steps in the well control procedure, and see what differences the
drilling method would require for implementation. Along the way the ambition was reduced
to focus on the analysis, as the development of a simulation model was proving too
challenging.
This work with this thesis would have been possible without the patience and guidance of
my supervisor John-Morten Godhavn. I am also very grateful for Espen Hauge’s help in my
attempt to model the drilling system described herein. Furthermore I appreciate the
opportunity to visit AGR’s workshop and office in Bergen, as well as their help, through Nils
Lennart Rolland and Kjell-Rune Toftevåg, and their willingness to support me with their
information.
I would also like to thank Statoil for the opportunity to do this thesis work with them, and for
supplying me with an office spot. During the report work, the online soruces such as
OnePetro supplied by NTNU has been most valuable. I would also like to extend my
appreciation to the open-source community of Zotero.org which supplies the reference tool
that has been invaluable to this work.
I hereby declare that the work in this project is made independently and in accordance to
the rules set down by "Examination regulations" at the Norwegian University of Science and
Technology (NTNU), Trondheim.
Trondheim, June 14th
2011
XMorten Hansen
Student
VIII
Table of Contents
List of tables ............................................................................................................................... X
List of figures .............................................................................................................................. X
1. Introduction ........................................................................................................................ 1
2. Dual gradient drilling ........................................................................................................... 2
2.1. Controlled mud pressure drilling system .................................................................... 4
2.2. Equipment ................................................................................................................... 4
2.2.1. Subsea pump module ........................................................................................... 4
2.2.2. Mud return line and handling platform ............................................................... 5
2.2.3. Control unit and umbilical .................................................................................... 5
2.2.4. Drillstring valve ..................................................................................................... 7
2.2.5. Coriolis flow meter ............................................................................................... 8
2.3. Possible benefits .......................................................................................................... 9
2.4. Challenges .................................................................................................................. 10
3. Conventional well control ................................................................................................. 12
3.1.1. Driller’s method .................................................................................................. 13
3.1.2. Engineer’s method ............................................................................................. 15
4. Automated well control .................................................................................................... 17
4.1. Backpressure MPD system ........................................................................................ 17
4.2. @Balance ................................................................................................................... 18
4.3. System value .............................................................................................................. 19
4.4. Micro-Flux Control ..................................................................................................... 20
5. Analysis of CMP well control procedures ......................................................................... 22
5.1. Normal Procedure ..................................................................................................... 24
5.2. Contingency procedure ............................................................................................. 33
5.2.1. Freefalling rate ................................................................................................... 33
IX
6. Discussion .......................................................................................................................... 44
6.1. The CMP technology .................................................................................................. 44
6.2. Automated well control ............................................................................................. 45
6.3. CMP well control ....................................................................................................... 47
7. Conclusions ....................................................................................................................... 50
7.1. Further work .............................................................................................................. 51
8. Nomenclature ................................................................................................................... 52
9. ............................................................................................................................................... 52
10. References ..................................................................................................................... 53
11. Appendix ........................................................................................................................ 56
X
List of tables
Table 1: CMP Equipment Dimensions and Sizes ........................................................................ 4
Table 2: Causes for kick (Grace et al. 2003) ............................................................................. 12
Table 3: Kick warnings (Grace et al. 2003) ............................................................................... 12
Table 4: Preparation for kick handling (Grace et al. 2003) ...................................................... 13
Table 5: Calculation parameters .............................................................................................. 34
List of figures
Figure 1: Difference between dual density gradients and conventional gradients (Rehm et al.
2008) ........................................................................................................................................... 3
Figure 2: DGD pressures and gradients vs conventional (Forrest, Bailey, and Hannegan 2001)
.................................................................................................................................................... 3
Figure 3: MRL hose ..................................................................................................................... 5
Figure 4: CMP system layout ...................................................................................................... 6
Figure 5: Anti u-tube valve, open position to the left and closed to the right (Rehm et al.
2008) ........................................................................................................................................... 8
Figure 6: Coriolis meter(Reitsma 2010) ..................................................................................... 9
Figure 7: DAPC system setup (Reitsma 2010) .......................................................................... 20
Figure 8:MFC system setup (Santos et al. 2007) ...................................................................... 21
Figure 9: BOP stack sketch ....................................................................................................... 23
Figure 10 Freefall rate vs. mud weight, pipe ID 1.75” ............................................................. 35
Figure 11: Freefall rate vs. mud weight, pipe ID 3.75” ............................................................ 37
1
1. Introduction
The aim of this thesis work has been to critically analyze the issue of well control using dual
gradient (DGD) systems, specifically AGR’s CMP system. In addition, the benefits and
challenges other than well control have been reviewed and discussed.
Furthermore it was necessary to review conventional well control, and also look into
automated well control systems offered for conventional drilling, and the potential they
might have for the DGD systems.
The work is based on discussion and analysis of the provided information from AGR
regarding their procedures for well control for their CMP system. Also Statoil’s internal
procedures and requirements for well control were used in the evaluation. These are
proprietary resources that have not been added to appendices or reproduced in any other
way than what is presented herein.
2
2. Dual gradient drilling
The aim with DGD is to allow the drilling of a well using heavy engineered mud, but still
keeping the gradient within the drilling window. This can be especially hard to achieve with
conventional methods when drilling through depleted zones, or in deep water areas.
This challenge can be met using a DGD system. The system uses two fluid columns, where
one is lighter than the engineered mud and thence modifies the gradient of the well. This is
generally done either by having no riser at all or to have the riser filled with a blanket fluid.
The mud returns are taken through a diverter from the wellhead or somewhere on the riser.
This is much like tophole drilling, just that the mud is in the hole and taken to surface
through a pump and mudreturn line (MRL). The system this report focuses on has in
addition to the mud return from subsea a riser. This allows for the interface between the
mud and the lighter fluid to be adjusted within the riser.
The subsea wellhead diverts the mud to a subsea pump that brings the mud to surface
through the MRL. The riser is then kept out of the circulation, and can be filled with a
blanket fluid. This is typically water, base oil or an inert gas like nitrogen. Different
technology suppliers generally keep a system of their own, and the blanket fluid typically
separates them. For instance AGR’s CMP system, which is the focus of this report, uses
water while ORS put their trust in nitrogen.
The result is that the gradients of the well are shifted. Both the mud and the well pressures
share the same overburden down to the mudline. This widens the drilling window as is
illustrated in Figure 1. In the same way the gradient of the mud is also altered. Since the
ratio between the light and heavy fluid will generally increase with depth, the gradient for
the well fluid will be dipping towards a vertical line as plotted gradient against depth, as
illustrated in Figure 2. This is the nature of the drilling method that gives the advantage of
choosing to skip a casing section if the pressures are right.
3
Figure 1: Difference between dual density gradients and conventional gradients (Rehm et al. 2008)
Figure 2: DGD pressures and gradients vs conventional (Forrest, Bailey, and Hannegan 2001)
4
2.1. Controlled mud pressure drilling system
CMP is AGR’s system for DGD with a riser. In their portfolio of drilling systems, it aims to be
a tool to improve drilling in the middle section, and together with EC-drill, the reservoir
section. In addition it enables drilling of tight windows not drillable with conventional
technologies. To date it has not been used on offshore wells, but the technology and
method is promising. Although the system is quite like with the RMR system, drilling with
CMP is all the more different in terms of applying the riser and the nature of the depths
drilled, with the possibilities of well control situations
2.2. Equipment
Description Dimension (m) [L x W x H]
Weight (Kg)
Riser Docking Arrangement TBD TBD Subsea Pump Module (3 -stage) 4.50 x 2.50 x 1.50 8500 Mud Return Hose 6" ID 12 /m Hose Handling Platform 3.10 x 1.84 x 2.72 3500 Umbilical TBD TBD Umbilical Winch (10 -ton SWL) 1 4.30 x 2.45 x 2.97 22000 CMP® Office/Tool Container 4.28 x 2.44 x 2.79 9000 CMP® Control Container 4.28 x 2.44 x 2.74 14000 Generator (Aggreko 1250 kW) 6.18 x 2.44 x 2.60 22000
Table 1: CMP Equipment Dimensions and Sizes
In Table 1 the equipment needed for operating AGR’s CMP system is shown, with specified
dimensions. Also in Figure 4 the system layout can be seen. The riser docking arrangement is
yet to be described in detail, but will be a docking above the subsea wellhead and BOP and
include the diverter for the subsea mud pump as well as the name states, a docking for the
riser to surface. The module along with the connection to the pump module will also have
to provide connection of choke lines to pump module, since the MRL and subsea mudpumps
need access to the well annulus during a shut-in and well control situation.
2.2.1. Subsea pump module
The subsea pump module (SPM) is highly important and the module that sets the limits of
the CMP operation. The pump’s lifting capacity dictates at what sea depth the connection of
the mud return line (MRL) will be, given a desired flow rate, or the other way around. The
pumps are Discflo, same as AGR uses in the RMR system (Frøyen et al. 2006) pumps, which
1 With platform and A-frame retracted, includes Umbilical on drum – Rated lifting capacity of winch frame.
5
are specially designed by AGR to allow pumping of mud with significant solids without
jamming or destroying the pump disks. The pumps are made up in the module 3 together
which one designated motor each.
2.2.2. Mud return line and handling platform
Another challenging piece of equipment is MRL and the handling platform. The length of the
hose, the weight of the mud it is to carry, and also the fact that it will be suspended and
flexible demands some expert engineering. The hose is supplied in 20 m lengths, and
requires a ¼ turn to make up, and also features external tension cables to further support
the weight of the massive MRL all the way to the SPM, see Figure 3. This hose is made of
rubber, although suggestions by AGR might indicate an alternative steel MRL is being
considered. The handling platform features a hang-off point where the lengths of MRL can
be made up to the line already in the water.
Figure 3: MRL hose
2.2.3. Control unit and umbilical
Then it is the control and operating modules, which are typical two standard containers,
holding the offices and control stations together with tools etc. The umbilical mentioned
above is the communication line between the subsea modules and the control and
operating modules, but also works as the lowering mechanism of the SPM, and is operated
6
through the umbilical winch. Other equipment needed includes also the flow return line,
which takes the mud return from the MRL to the rig mud system, through a manifold and
also a Coriolis flow meter and automated choke setup. Both of these are used to handle a
possible well control incident..
Figure 4: CMP system layout
7
2.2.4. Drillstring valve
The drillstring valve is designed to arrest the u-tube effect. The anti u-tube valve is essential
in the procedures for well control with CMP. In some cases it is necessary for the usage of
CMP throughout the well-control. In the case of no or a malfunctioning anti u-tube valve the
procedures might be totally altered, or even conventional procedures have to be used. The
valve was designed and built to hold the hydrostatic column of the whole
drillstring.(Gonzalez and Smits 2002; Gonzalez 2001)
The valve will normally be adjusted for every section of the well, or when mud weight is
changed. Thus it can be set to hold the hydrostatic column of the mud inside the drillstring
to the depth and end of each section. The valve will then open at a set pressure, and allow
flow through it. This complicates slightly hydraulic calculations, and also adds slightly to
frictional pressure drops in the drillstring. However, the purpose to arrest the u-tube effect
overcomes these slight downsides. Without it, the u-tube effect will cause the mud in the
drillstring to freefall unless the rate is high enough, and thus increase the bottomhole
pressure (BHP) and possibly fracture the well.
The valve is designed with springs that hold seals in place. See Figure 5 for the basic design
of the valve, with the open position to the left and open to the right. The springs can be
adjusted to the desired opening pressure, such that it fits to the mud and depth of each
section. This adjustment can only be made while the valve is out of the hole though.
Oskarsen (2001) did studies on the hydraulics of the valve, and also analyzed the location of
the valve in the drillstring. He arguments that the hydraulics would require it to be as close
to the mudline as possible to allow lower circulation rates to open it. Although this will still
arrest the u-tube and work as intended, it would not allow for the use of any wireline should
it be necessary. More importantly, it would hinder communication with the BHA whenever
it is closed. That leads to the placement of the valve to be below the MWD tools so that
communication with these tools will not be hindered.
8
Figure 5: Anti u-tube valve, open position to the left and closed to the right (Rehm et al. 2008)
2.2.5. Coriolis flow meter
The Coriolis flow meter is an accurate flow meter that applies the Coriolis effect to measure
flow. The Coriolis effect is shortly described the effect where a fluid flow in a rotating
reference frame is deflected. The effect is generally experienced for example in a sink:
When water drains in the sink, it will whirl in one direction on the northern hemisphere, and
another in the southern hemisphere. This effect is caused by the earth’s rotation, which is
the rotating reference frame.
The Coriolis meter applies this in a looped setup, where the flow runs through a looped
setup, and sensors register how much the flow changes the vibration of the tube. Figure 6
shows the setup of the meter. The physics of the Coriolis effect show that the acceleration
caused by the flow is proportional to the mass flow. The meter can also be set to measure
the density of the flow, and therefore accurately measure the volume flow as well as the
mass flow. The meter is accurate to 0.2 percent (White 2008)
9
Figure 6: Coriolis meter(Reitsma 2010)
2.3. Possible benefits
The benefits from DGD are mainly economical, and possibly also safety-related. This
however requires sound testing and analysis to state for a fact. However, it is not difficult to
comprehend that using a riser filled with seawater may save significant amounts of mud,
and may also require less storage capacity for mud on the rig, especially in long and
deepwater wells. Even more mud may be saved by the fact that the probability of lost
circulation is lessened, and if it should happen, the BHP can be very quickly adjusted.
Compared to conventional drilling where new mud has to be weighted and circulated, an
adjustment in subsea pump speed and flow may adjust the BHP much more efficiently and
faster. The greatest pressure properties of the DGD system is illustrated very well in
previously mentioned Figure 1 and Figure 2, showing how DGD gradients better adapt in
narrow windows and in long wells, eliminating casing sections.
The similarities with RMR are many, even though they are quite different in some ways.
Some of the benefits experienced with RMR on live wells can also apply to CMP. According
to R. Stave et al. (2005) the benefits may also include: Improved well control, reduced
numbers of casing and liner strings, reduced chances of loss of circulation. It is also
10
interesting to note that with RMR Stave et al. also states that aside from the rig-up, the
usage of the extra equipment doesn’t require additional time compared to conventional
use. It would be reasonable to think this would go for CMP as well, as the only other
equipment affecting operations is the addition of the riser instead of the suction module.
When used on floaters, the CMP system can prove beneficial when the need to quickly
disconnect arises. As the riser will only be filled with blanket fluid or seawater,
disconnecting should not affect the pressure control of the well, since no fluid column is
lost. The pump can quickly adjust the pressure to ensure the safety of the well before the
BOP is closed and the disconnection is completed. There is then also limited spill, as the
volume of the hose will be significantly smaller than the volume of the riser.
2.4. Challenges
According to AGR one of the challenges with the equipment has been the mud return line
(MRL), making it both big enough to allow for flow, but small enough so it still is flexible and
easy to use. The mud it carries makes up the biggest load on the hose, and the loads on the
hose are significant, and this is the major challenge while trying to find the right size for it.
Thus it has also been proposed to use steel pipes to increase the strength, but this would
also mean the weight of the MRL will increase. However, for the current purposes the
designed hose works fine, as the pumps restrict the possible water depth it can operate, and
the hose is designed with this in mind. It has been mentioned that the limit is around 500m
of water depth to the pump. This does not mean that the CMP system can’t be utilized in
deeper waters, just that the pump tie in to the riser will be in mid water. Then again,
investigations should be done to look at how the distance between a deepwater BOP stack
and the shallower pump module would affect well control and other operating challenges.
In case of equipment failure, the well will in many cases be returned to a conventional state.
In most cases, this doesn’t have to be critical. For instance if the mud pump fails, the returns
can be taken through the riser as conventionally. That will result in the loss of the CMP
advantages however, and in case the pressure profile of the well at that time requires CMP
flexibility in the BHP, the equipment failure can result in loss of circulation and following
problems.
11
The pump may be considered the piece of extra equipment compared to conventional
drilling that increases the risk and possibility of equipment failures. The BOP is essentially
the same, apart from the pump tie-in, the riser is the same and doesn’t see the same
pressures. The MRL and flow-return topside can essentially be viewed in the same manner
as the rig mud handling system. However, there are still the tension problems that can be
associated with the hose. On floaters the tension problems can be worse due to heave and
rig movement in relevance to the wellhead.
Well-control with CMP will be reviewed in the following chapter 5.
12
3. Conventional well control
The general drilling method has remained the same for decades, and the same goes for
well-control. The methods and procedures are well established, yet there remain two
different methods of handling a kick. They are called the Driller’s method and the Engineer’s
method (Wait & weight method) referring to the latter as more calculation intensive, hence
the name. A brief description of these established well control procedures are included as a
base-reference for the evaluation of the CMP well control procedures. As should be noted,
no loss of safety should occur when implementing new technologies, therefore the CMP
procedures should maintain even better well control than these conventional ones.Grace et
al. (2003) lists the causes of kick in Table 2, and the indications of a kick in Table 3.
1. Mud weight less than formation pore pressure
2. Failure to keep the hole full while tripping
3. Swabbing while tripping
4. Lost circulation
5. Mud cut by gas, water, or oil”
Table 2: Causes for kick (Grace et al. 2003)
1. Sudden increase in drilling rate
2. Increase in fluid volume at the surface, which is commonly
termed a pit level increase or an increase in flow rate
3. Change in pump pressure
4. Reduction in drillpipe weight
5. Gas, oil, or water-cut mud
Table 3: Kick warnings (Grace et al. 2003)
13
When a kick has been identified the normal shut-in procedure shown in Table 4, is used
regardless of which circulation method is chosen (Driller’s or Engineer’s).
1. Drill no more than 3 feet of any drilling break.
2. Pick up off bottom, space out, and shut off the pump.
3. Check for flow.
4. If flow is observed, shut in the well by opening the choke
line, closing the pipe rams, and closing the choke, pressure
permitting.
5. Record the pit volume increase, drillpipe, pressure, and
annulus pressure. Monitor and record the drillpipe and
annular pressures at 15-minute intervals.
6. Close annular preventer; open pipe rams.
7. Prepare to displace the kick.
Table 4: Preparation for kick handling (Grace et al. 2003)
3.1.1. Driller’s method
The Driller’s method is the most widely used and the standard method to circulate a kick out
of a well. Its main advantage is that it is simple and requires minimal calculations. In
simplicity the Driller’s method circulates the kick in one circulation, while the second
circulation effectively kills the well. One circulation meaning pumping once the whole
volume of the well, also referred to as bottoms up (BU). This is done to try to keep the
bottomhole pressure stable and ideally constant. However, this procedure might be more
time consuming, and applies a higher pressure to the well.
14
The Driller’s method can be summarized in the following steps
1. Close in the well
- Raise the top drive
- Stop the pumps
- Close preventer or pipe ram
- Open choke line valve against closed choke
2. After sufficient time for the first pressure build up, pump slowly to open the float
and read shut in drillpipe pressure (SIDPP)
3. Circulate with predetermined kill rate, when keeping choke-pressure constant. Read
standpipe pressure and calculate.
4. Circulate out the influx at a predetermined pump rate and control the choke to keep
the standpipe pressure constant. Make sure that all influx is circulated out of the
well before proceeding to the next step.
5. Displace the existing mud using a predetermined pump rate and a new mud
weighted to balance the formation pressure, thus killing the well.
6. Control the choke to give constant standpipe pressure while circulating the new mud
up the annulus.
The Driller’s method can also be summarized in a killsheet as prepared by IADC, please see
Appendix C.
15
3.1.2. Engineer’s method
The Engineer’s method, also referred to as the wait and weight method, is a more
calculation intensive method than the driller’s method. It only applies one BU to both kill
the well and circulate the influx out. This is done by waiting for the weighting of kill mud,
and circulating the influx using the kill mud.
The Engineer’s method can be summarized in the following steps:
1. Close in the well
- Raise top drive
- Stop pumps
- Close preventer or pipe ram
- Open choke line valve against closed choke
2. After sufficient time for the first pressure build up
- Read shut in annulus pressure (SICP). Pump slowly to open the float and read
shut in drillpipe pressure (SIDPP)
- Determine surface volume gain
- Determine the new mud weight for killing the well, SGkill. Use the true
vertical depth (TVD) of the bit in a deviated well.
- Determine preparations for killing, watch drillpipe pressure (SIDPP) and
casing pressure (SICP) for a second pressure build-up, resulting from gas
percolating up the annulus. If necessary, bleed off mud from the annulus
until drillpipe pressure reverts to the original shut in drillpipe pressure
(SIDPP).
3. Start killing with the new mud weight as soon as possible in the following way:
- Open choke and start the pumps at a pre-selected reduced killing speed
16
- Regulate the choke opening to the initial choke pressure (SICP). Read the
corresponding drillpipe pressure which will be initial circulating pressure
(ICP).
- Find friction loss in system= ICP – SIDPP
- Determine the surface to bit pump strokes and travel time (t1) and bit to
surface pump strokes and travel time (t2)
- Calculate either the new friction loss(Pc2) from the previous (Pc1) using Eq
3.1 or the final circulating pressure(FCP) from the pressure at slow circulating
rate(SCR) using Eq.3.2. SGkill is the kill mud weight and SG1 the original mud
weight
��� � ��� · ������� (3.1)
� � � � � · ������� (3.2)
- Regulate the choke opening during t1 in such way that ICP drops gradually to
FCP while maintaining the same pump speed.
- At the end of the interval t1 when the new heavy mud (SGkill) has reached
the bit, continue to circulate the influx out of the hole. Maintain pump
strokes constant and regulate the choke opening in such a way that the FCP is
the observed standpipe pressure.
The Engineer’s method can also be summarized in a killsheets as prepared by IADC, please
see appendices A and B.
17
4. Automated well control
There are suppliers that have developed systems to handle kick detection and controlling
the handling of the kick. These systems can be applied to both MPD and conventional
drilling, although the application will most likely be with MPD systems, where there already
are lot of the systems and equipments used in place.
With the general development of technology, automation is also coming to the drilling rig.
With more advanced flow and pressure meters, the potential for more accurate handling of
well control is apparent. In the first instance it is the detection of kick that is being
upgraded, with hydraulic models monitoring the measurements from flow meters, pressure
transducers and rig pumps. In some cases measurements from MWD tools and especially
pressure while drilling (PWD) data further improve the potential for well control helped by
automated systems.
There are two main competitors in this field. @Balance offers Schlumberger’s project
Dynamic annular pressure control (DAPC), and Secure Drilling is a part of Weatherford. Both
are offered as enhancements to conventional drilling, but are for the most part applied as
integrated parts of various MPD systems. The knowledge of the systems is limited to
published materials, so now thorough procedures can be reviewed, only the general
concepts of the technologies.
4.1. Backpressure MPD system
Since both the automated well control systems reviewed are developed on the basis of a
backpressure MPD system, here follows a short description of the basics of drilling with a
backpressure system.
By drilling underbalanced, but by adding a backpressure, the resulting ECD is in balance.
While making connections, the driller can maintain the bottomhole pressure by applying a
backpressure and sealing the wellbore with a continuous circulation device, like a
continuous circulation coupler (CCS) or a rotating circulation device (RCD). The backpressure
is added in either one of the two methods:
• Can be achieved in an open system by using a continuous circulation method
18
• Can be achieved in a closed system by using a backpressure setup
The system seals off the wellhead and applies a surface backpressure by restricting the
return flow through a choke. Thus the backpressure can replace the loss of pressure when
the rig pumps are turned off, and causing the loss of frictional pressure. This backpressure
can be applied both when drilling, and during connections.
This gives us more control of both static and dynamic pressure, and although it might not
always be possible or necessary to make them equal, it enables us to better control the
pressure within a pressure range defined by downhole conditions.
The control of this system can also be achieved by an additional pump at the backpressure
line, increasing control, magnitude and precision of the applied backpressure. The system
can also be run automatically through a control system which applies a hydraulic model,
downhole pressure data from pressure while drilling (PWD) tools, and may also include
flowmeters to enhance kick detection.
By using a back-pressure pump, the system can exercise an even faster response to pressure
changes in the well. In many cases, a choke system will not be fast enough if something
dramatic happens to the BHP like a pump failure or a human error. Also, a choke can only
trap the pressure that is left in the well when it finally is able to close. The back-pressure
pump can respond quicker, and can apply more pressure than a flow restriction can. The
precision of back-pressure from a pump is also greater at lower flowrates compared to
choke.
4.2. @Balance
The systems provided by Schlumberger are offered through @Balance, and includes several
systems related to MPD and UBO drilling. The system for kick detection and pressure control
is referred to as Dynamic Annular Pressure Control (DAPC).
The DAPC is a system designed as a part of a backpressure MPD system. This system applies
a backpressure pump to keep BHP constant when not circulating, most typically during a
connection. The system also involves an automated choke, and the control system. An
illustration showing the setup of the system can be seen in Figure 7. The kick detection has
been done in various ways. Fredericks et al. (2008) applies a Coriolis flow meter, while later
19
Reitsma (2010) arguments that the system can detect kicks better by applying a model that
looks at trending data for annular discharge pressure and standpipe pressure.
In the full scale test described by Fredericks et al. the system is a part of several other
drilling system providers. The DAPC is handling kick detection and control, and analyzing the
hydraulics. In this case wired drillpipe and real time PWD measurements were used to
calibrate the hydraulic model while drilling ahead. This case thus applies several high end
technologies to further the well control, as it was described a prerequisite for the project
which Fredericks et al. is referring to.
The system is applied in this instance to one of the more widely used MPD methods, as
backpressure control is a highly effective way to control BHP without adding or modifying
too much equipment. The system should therefore be thoroughly tested and be well
developed. By looking at the work done by Reitsma (2011) it could be indicated that the
system is being developed further, or that the kick detection with Coriolis meter isn’t
working optimally. However, this shows that the system has been successfully tested and is
still in development. Further full scale test results should be published to prove the value of
the system
4.3. System value
As it is offered as a system that can be implemented to other drilling systems together with
other suppliers, this could prove a promising system. However, such combinations could
also present major challenges when the interfaces of different systems don’t always work
well together. The tests done by Fredericks et al. seemed positive, but other full scale test
results hasn’t been published since, so it can’t be made for fact that the system is fully
operational. This could be because the focus of the project is on the system handling the
hydraulics, and that to apply it to drilling it needs to be combined with an advanced MPD
system. Thus there might be the challenges of competing companies and other
management related problems. Then again, the core of such a system is a hydraulics model
and programmed controllers for chokes and pumps. Those aren’t proprietary and very high
end developments, and it might be that suppliers of MPD systems can well enough develop
their own systems for this.
20
Figure 7: DAPC system setup (Reitsma 2010)
Should @Balance supply the whole MPD drilling package however, the experience with the
system in operation, and a well built model with a lot of data available for calibration could
prove to be a great advantage.
4.4. Micro-Flux Control
Micro-Flux Control (MFC) is Weatherford’s system for pressure control, developed by Secure
Drilling. The system aims to supply increased safety and reduce risk of kicks with all drilling
systems on all rigs, according to (H. Santos, Reid, and Lage 2004).
The system is also a backpressure MPD system that applies rigid pressure control to a
closed-loop drilling fluid system. As with the above mentioned @Balance DAPC system, this
is based on the MPD backpressure system, but may be applied to other drilling systems as
well. Through controlling a backpressure pump and an automated choke, together with a
RCD completing the closed circulation system, the system setup is shown in Figure 8. Similar
systems are offered by several others, but the thing separating the MFC system is that it
applies such strict control on the flow measurements and well control that it can handle
pressure tests while drilling ahead (H. Santos, Reid, and Lage 2004).
21
These pressure tests will be used to calibrate the pressure control of the well as drilling
progresses. Whereas the DAPC of @Balance will calibrate its hydraulic model real time with
hydraulic and PWD data from the well, the MFC system will measure fracture and pore
pressure as it drills ahead. Since the system is closed and the sensitivity of the flow
measurements are so good, the risk of what is referred to as “micro influxes” or “micro out
fluxes” by the authors, it can take leak-off and inflow tests as it drills ahead. However, (Helio
Santos, Leuchtenberg, and Shayegi 2003a; 2003b) don’t state what method is used to
ensure such accurate flow measurements. However, from the schematic in Figure 8, it may
be glanced that item 2 looks very much like a Coriolis flow meter. Item 1 would be the RCD,
and item 3 reflects that the control system is the important third part of the system.
Figure 8:MFC system setup (Santos et al. 2007)
22
5. Analysis of CMP well control procedures
The well control procedures for CMP has been put down by AGR in a slideshow depicting
the well and the fluid system for several steps throughout the process of handling the kick,
as well as an official procedure plan for the incident. Both of these have been used in the
following analysis of well control with CMP compared to conventional well control. Figure 9
shows the setup of the subsea BOP stack used for CMP drilling, and is essential in this
analysis. The well control procedure steps as shown requires a working anti-U-tube valve in
the drillstring. See chapter 5.2 for alternative procedure for drillstrings with nonexistent or
malfunctioning anti-U-tube valve.
23
Figure 9: BOP stack sketch
24
5.1. Normal Procedure
The procedure denoted as normal procedure is the well control procedure for a drillstring
with a working anti u-tube valve. For contingency or alternative procedure, it is referred to
the following chapter 5.2. The numberings below each step headline are referring to the
numbered steps in the slideshow (SS) showing the steps, and the procedure document (D)
both provided by AGR.
Step 1: Detection
(SS: 1-5, D1)
Detection of the kick is the first step in any well-control procedure, and for conventional
drilling it is mentioned above that pit gain and monitoring the trip tank is the main
indication of a kick. For CMP the pit gain comes in addition to some more advanced
indicators. The CMP pumps will show an increase in speed\power usage to compensate for
the influx. If the pump is set to keep a constant suction pressure, this will result in the speed
of the pump to increase. If the pump is set to keep a constant speed, the pressures will
increase indicated by the pressure transducers. Furthermore there will be a Coriolis flow
meter on the mud return manifold topside that will further help identify an increase in flow
return. The principle the meter is based upon indicates a flow increase much more precisely
than a pit gain may. Thus, influx volumes down to half a barrel may be identified.
Comment
It seems that the detection of the kick is more rapid with CMP, having both the flowmeter
and pump indicating an increased flow in addition to the traditional pit gain. However, the
sensitivity of half a barrel may seem unnecessary, as I assume there will never be such a
stable flow return that once there is a deviation of half a barrel, we can assume it’s a kick.
Care should be taken not to too quickly flag a change in flow returns as a kick. Furthermore,
with the u-tube phenomenon, even with an anti-u-tube valve, there may come air into the
drillstring. This is a result of the mud freefalling, and this will happen even with the arrestor
valve, below it. Then small bubbles of air come into the well, and passing the subsea pump
they will cause it to operate abnormal and potentially be flagged as a kick. These problems
will need to be mitigated with sound procedures and limits, to avoid unnecessary warnings.
This is not mentioned in the procedures, either in the document or the presentation, but it
25
should be implemented in the operating software. However, a written procedure should be
prepared, so that this can be manually monitored and verified for contingency.
Step 2: Shut-in
(SS 6-7, D 2-7)
Stop drilling and space out drillpipe, prepare to shut in well, stop rig pumps, stop CMP
pumps. It is important that the rig pumps are stopped before the CMP pump so that the
mud level doesn’t start to travel up the riser. The valve on the MRL should close
automatically to prevent u-tubing into the riser. The BOP and CMP pressures should be
monitored to verify kick, and try to estimate the magnitude. Then the annular preventer is
closed
Comment
The shut in of the well following these steps is known as a soft shut in, done to mitigate u-
tubing problems. This differs from normal shut-in procedures where hard shut ins, with
annular preventer being closed before pumps are shut off are normal on the Norwegian
continental shelf (NCS).
Step 3: Stabilize and monitor well
(SS 8-9, D 8-9)
The well is monitored and stabilized, as with any well control method. The kill calculations
are done and a kill schedule is made.
Comment
This step is the same as with any well control procedure, although the calculations needed
are different since the system is different.
Step 4: Close isolation valves
(SS 10A, D 10)
A number of isolation valves are closed to prepare the system for handling the influx.
Referring to Figure 9, Valve 1 is closed and isolates the main suction line from the subsea
pump. Valve 2 is also closed, isolating the choke-line and redirecting it into the mud pump.
26
Comment
With Valve 1 closed the riser is isolated and the choke is lined up and ready to be handled
through the CMP system. This is only a number of steps where valves are used to divert the
flow from the riser to the CMP and MRL system. It is worth to note that for every valve used,
there is an increased chance of failure. The valves are failsafe closed, so the contingency is
that should any of the valves fail, it is still possible to handle the kick conventionally through
the riser.
Step 5: Open choke line to CMP
(SS 10B, D 10)
Again referring to Figure 9, Valve 3 is to be opened and Valve 4 closed. Now the BOP is
controlling the choke line valve (Valve 4), and the CMP is controlling the isolation valve
(Valve 4).
Comment
The CMP now can actively start handling the kick through the choke line and subsea pump.
Both the subsea pump and the isolation valve mentioned may help control the pressure and
flow from the well. This is also where the well control now will be different from
conventional control, as the flow out of the well is controlled by both a choke line valve at
the subsea head, as well as the subsea pump. The equipment handling the kick here now will
be bigger, as a MRL is much larger than the choke line, and the pump also is a difference.
Step 6: Obtain pressure readings and start rig pumps
(SS 11-15, D 11-15)
This assumes a working anti U-tube valve as described is part of the bottomhole assembly
(BHA). The rig pumps will be started slowly to obtain the pressure required for opening the
anti u-tube valve.
The rig pumps are started at slow rates and slowly increased, to monitor standpipe pressure
(SISPP) while building up pressure to open anti u-tube valve. The valve is predesigned to
open at a pressure specific for the section currently being drilled. The opening pressure of
the valve should also have been fingerprinted throughout the section. Thus, when the valve
27
opens, that’s the maximum SISPP. Now the kick pressure can be calculated using eq 5.1 and
pumps are shut off again. Here �� is the kick pressure, and ����� is the opening pressure
of the anti u-tube valve.
�� � ����� � ����� (5.1)
Comment
This step is done to obtain the kick pressure, and thus allowing the correction of the suction
pressure as in the next step. Doing this isn’t out of the ordinary; it would have had to be
done in conventional operations as well if there was a float valve in the drillstring. The
fingerprinting is also done in conventional operations, so in general this step shouldn’t be out
of the ordinary.
Step 7: Adjust suction pressure and decide circulation method
(SS 16, 17, D 16-18)
The kick pressure calculated in the previous step is added to the pump suction pressure, in
addition to a safety margin typical of 50 psi, but this may be adjusted to fracture pressure.
This suction pressure should now help keep the kill rate throughout the well. Using eq. 5.2
we can now calculate the new set suction pressure. Here ���� is the new set suction
pressure, while ����,����� is the original set suction pressure, and ������� is the safety
margin that applies to the well.
���� � ����,����� � �� � ������� (5.2)
Comment
The suction pressure set this way ensures a uniform flow throughout the well. The difference
from conventional method here is that the kill rate needs to be controlled in “both” ends of
the well, so as not to cause the well to build up fluid and pressure, or lower it. However it is
not mentioned here, it is assumed that all the same, care should be taken to set the suction
pressure based on PVT analysis as well. The pump will not work effectively should it pump
gas, so the suction pressure should ensure that no gas boils out prior to passing the pump.
This is of course dependant on the well. Furthermore it is a lower limit to the maximum
possible pressure the CMP system can handle compared to a conventional system handling a
28
kick. Furthermore the pressure limits of the CMP equipment must be considered. Should the
casing pressure exceed the operating limits of the CMP equipment, the kick will have to be
either circulated conventionally through the choke line, or be bullheaded.
Step 8: Start CMP pump
(SS 18-20, D 19-20)
The CMP pump is restarted with the new suction pressure set in the previous step. This
should be close to the casing pressure. Valve 4 is opened and the CMP pump now
communicates with the well below the annular preventer through the choke line. The set
pressure is only a starting point, and pump control should be switched to set speed.
Comment
This is where the CMP actively start handling the kick, with the flow from the well now going
around the preventer through the chokeline to the CMP mud pump. It is interesting to not
that the CMP pump should be switched to set speed after the pressure has equalized to the
set suction pressure from step 7. This will allow the pressure just below the BOP to vary,
while the flow from the well is steady.
Step 9: Start rig pumps and switch flow return
(SS 21, D21-24)
The rig pumps are started and brought up to kill rate. Flow return switched from Coriolis
meter to choke and degasser line. Before switch, check that CMP flow return equals flow
rate from mud pumps.
Comment
Now the CMP pump will start automatically, as the pressure builds up after the rig pumps
are started. The monitoring of the CMP pump will be important here, to ensure that outflow
equals inflow to prevent pressure build up. This should take care of itself if the suction
pressure is set correctly. That will enable a switch to constant speed for the pump, so that
the flow rate becomes stable. This will be the most effective way to ensure a constant rate,
as the control of the pump doesn’t allow a set flow rate without doing some intermediate
calculations.
29
Step 10: Record pump speed and rates, switch CMP pump control to fixed speed
(SS 22-23, D 22-23)
Monitor the flow of the well and record pump info such as speed and rates, keep
outflow=inflow. When the flow is balanced, operate CMP at fixed speed.
Comment
This should be one of the advantages of CMP, where the flow is very well controlled, since
the CMP pump is actively controlling the outflow, and the Coriolis meter as mentioned
provides better flow reading than conventional methods. The control of the CMP pump, both
when reaching the flow balance, but also at fixed speed should be simple with the operating
software of CMP. However, it is important that if this is to be automated, it should still show
info so that this can be manually monitored all the time.
The document doesn’t specifically tell to monitor pump speed and rates after the CMP
control is switched to constant speed. However, it does mention to set the automated choke
to keep the standpipe pressure constant, so that also the BHP remains constant. This is now
where the complexity of the pressure situation in the well starts to show itself. Considering
rig pumps are held at the kill rate, there is still the control of the CMP pump and topside
choke valve that needs to be coordinated to control the pressure. The topside choke will
control the BHP. However, there are now a number of points of interest regarding pressure,
that all will be adjusted simultaneously: suction and outlet pressure of pump, BHP and the
point of boiling for the gas if applicable. Should the control system be inexact or
malfcuntioning, the satisfactory control of all of these points of interest could prove difficult.
30
Step 11: Maintaining constant standpipe pressure
(SS 24-26)
Flow on return manifold is switched to topside choke and degasser. The choke is automated
and the set point should be adjusted to keep the standpipe pressure constant.
Comment
This is another operation that can be automated, and if the equipment works the control of
the circulating kick is increased. If choke topside should fail, there is no barrier against
possible boiling of gas in MRL, still the same as with conventional. If automated, need the
possibility of quickly manually controlling it. This may require extra thought on how to place
the extra equipment of CMP such as the manifold and its choke.
Step 12: Circulate the kick out, monitor well
(SS 27,28)
While continuing to circulate the kick out of the well, keep monitoring the situation and
pumps, as well as standpipe and casing pressures. If some values seem unnatural, shut
down pumps and close the automated choke and diagnose the situation. When the kick
eventually boils out of solution, the pressure increase will cause the choke to close to keep
pressures constant.
Comment
The automated choke should in this phase of the operations control the well automatically,
helped by the operating system. This is the phase where the CMP automated system really
can come to its benefit, where the control while gas is boiling is good, as long as the
operating system and its models can handle it. As mentioned above, there will be an issue
here of trying to keep the kick from boiling before reaching the pump.
31
Step 13: Confirm kick has been circulated out of well
(SS 30, D30)
By inspection of the return flow and pit volumes, it should be agreed that the kick is out of
the well. Now the BOP stack has to be flushed and the pressure across the annular
preventer needs to be equalized.
Comment
Standard procedure
Step 14: Flush BOP stack and circulate out flushed gas
Use BOP control to open BOP valve and flush and circulate out CMP pump and MRL
Comment
Standard procedure
Step 15: Determine kick pressure for killing the well
( SS 33-36, D33-36)
Using the pressure transducer connected to the riser, the pressure difference over the
annular preventer can be determined. Thus the procedure to equalize the pressure can be
begun after mud return line and CMP pump are shut off. There are 2 methods to restore the
pressure across the preventer. They are described as Method A and Method B
Comment
This is when CMP might be a bit more complex than conventional drilling, as there are more
pressures to evaluate and a little more to calculate. Differential pressure over the annular
preventer will be the pressure difference between riser pressure and casing shut in pressure
Method A
32
This method is much the same as the second step in the conventional Driller’s method.
Heavy mud is used to displace the well below the annular preventer to kill the well and
equalize presser across BOP.
Step 1: Adjust mud
(SS 39,40, D 39, 40)
Looking at the suction pressure used, the delta pressure from this to the ambient water
pressure at the pump depth will be the increased well pressure, which the mud weight
should apply. Then the mud should be prepared to add this pressure to the well
Step 2: Circulate
(SS 41)
The heavier mud is pumped into the well, and CMP suction pressure should be reduced as
the well pressure increases. When the well is filled with the new, heavy mud, the pressure
across the BOP should be equalized. Should the riser be filled with other blanket fluid, this
should be adjusted for when determining the target suction pressure.
Comment
This is really just a change in the mud, but the calculations are a bit different from
conventional operations. Furthermore, the CMP system requires the suction pressure to be
adjusted at the same time, but the operating software should take care of this easily.
However, this requires the pressure transducer to be working, unless the well should fracture
and loss of circulation occur.
Method B
This method is quite simple. The riser is filled through a boost line with mud until the
column above the BOP equalizes the pressure across it.
Comment
This is an easy way to equalize the pressure, but has a few shortcomings. The riser needs to
be long/deep enough so that a proper mud column can be placed on top of the BOP.
33
Furthermore this goes against the CMP basics, with filling the riser, thus there might not be
enough mud on the rig to do this, or possibly the boost line might not work properly.
5.2. Contingency procedure
With no or malfunctioning procedure, the normal procedure cannot be used, thus this
contingency procedure to handle u-tubing has to be used. This procedure is only presented
in the procedure document, not the slideshow.
Step 1, Fingerprinting
(1)
Fingerprint the pump rates needed to keep 50 psi (3.4bar) on the standpipe. This is done to
keep the standpipe pressure above zero in the case of a well control situation, so that
standpipe pressure can be used to adjust choke pressure
Comment
This is keeping the mud from freefalling and creating vacuums in the drillpipe and thence
removing the standpipe pressure transducer and pump pressure communication with the
well. This might be a challenge though. From some simple calculations shown below it can
be anticipated that keeping the mud from freefalling might be a challenge. The
fingerprinting is also important with a functioning anti u-tube valve, to know the slow
circulation rates and frictional pressure losses. However, without the valve it is more so
important to ensure that the string is always filled up to the standpipe.
5.2.1. Freefalling rate
The phenomenon known as freefalling is when the mud inside the pipe will flow faster than
the pumps can keep up. In general the weight of the mud is greater than the friction force,
and the resulting flow rate is greater than the pump rate. Some vacuum mechanics might be
affecting this as well, but with some simple calculations the matter has been illustrated.
The API have declared formulas to be used for the calculation of frictional pressure losses,
for either drillpipe or annulus, and for the properties of the mud, either the Newtonian,
Bingham-Plastic or the Power-law method. (Bourgoyne 1986, 2:155) For these calculations a
lot of assumptions were made to simplify, and focus on the result; that high flow rates will
34
be needed to prevent freefalling. From the formulas declared by API and given by
Bourgoyne, the formulas for turbulent flow with Bingham-Plastic fluid were chosen. A
vertical well, as a simple case that induces the largest freefall rates was chosen.
∆���� � ∆� ��!�� � ∆������� (5.3)
∆������� � "#.%& · '(�.%& · )�#.�&1800 · -�.�& · ∆. (5.4)
∆���� � /0� � ��1�� (5.5)
∆� ��!�� � " · 2 · 3 (5.6)
∆� ��!�� � 0.052 · " · 3 (5.7)
Now since eq. 5.4 involves a constant and the API formulas uses field units, we can rewrite
eq 5.6 to 5.7
By inserting eq 5.4, 5.5 and 5.7 into 5.3, and order it to express the mean velocity, we get eq
5.8
'( � 67/0� � 0.052 · " · 38 · 1800 · -�.�&"#.%& · )�#.�& · ∆.9.:;
(5.8)
Now since the eq xxx involves a constant and the API formulas uses field units, we can
Surface pressure 50 psi 3 bar BHP 5000 psi 345 bar Length 7000 ft 2134 m
Table 5: Calculation parameters
The pressures and well depth was chosen randomly, as emphasis is put on mud and pipe
variations and their effect on the freefall rate.
To illustrate the variation in freefall rates with mud, the freefall rate was plotted against the
mud weight, with a given pipe ID and mud rheology. For pipe with 1.75” id, the freefall rate
35
is plotted in Feil! Fant ikke referansekilden. with one graph for mud with plastic viscosity of
25 and 40 cp, and the same in Feil! Fant ikke referansekilden. when the pipe id is 3.75”.
From Figure 10 it is apparent that there is a difference in the rate to keep up with freefalling
from the two fluid viscosities. What the graph doesn’t show is the domain between 25 and
40 cp for mud with the same weight. However, it is clear that the rheology can affect the
freefall rate as much as the mud weight. At the same time, rheology has a direct impact on
the freefall rate, while the effect mud weight has can be more or less depending on BHP and
surface pressure. The rheology effect will also be affected by depth, as the friction will
increase with the length of the string. However, the graph shows that in general, for the
same mud weight, the higher the viscosity the lower the freefall rate. The effect is here
about 350 LPM (liters per minute) between 25 and 40 cp. The outer points of weight, with
1.20 and 2.20 is roughly 400 LPM.
4 700,00
4 800,00
4 900,00
5 000,00
5 100,00
5 200,00
5 300,00
5 400,00
5 500,00
5 600,00
5 700,00
1,00 1,20 1,40 1,60 1,80 2,00 2,20 2,40
Pu
mp
ra
te [
LPM
]
Mud weight [SG]
Pipe ID 1,75"
Cp=25
Cp=40
Figure 10 Freefall rate vs. mud weight, pipe ID 1.75”
36
Looking at Figure 11Feil! Fant ikke referansekilden. it is apparent that the rates are much
higher with pipe ID 3.75” than with 1.75”. These pipe sizes are not meant to refer to specific
drillpipes, just used to illustrate the effect pipe ID has on the freefall rate. The freefall rate is
here over 40 000 lpm for some mud weights. This might show that the calculations are
simplified just a bit too much, and that more thorough analysis should be done. On the
other hand, it is beyond doubt that the rate to keep up with freefalling mud will be high.
Drillpipes with ID around 1.75” is 2” OD, a size that is rarely used on the NCS. With this pipe
size requiring around 4000 LPM in these calculations, there is reason to believe that for
normal used pipes with ODs from 5” to 6 5/8” will have significant flow rate requirements if
freefalling is to be avoided.
Again, these are simple calculations, with quite arbitrary picked parameters, to show the
magnitude of flow rates needed, as well as the impact mud weight, rheology and pipe size
has on the freefall rate. More thorough simulations with complete well set up, as well as
flow up the annulus should be looked into to further determine the severity of the freefall
rate problem.
37
Step 2: Detect influx
(2)
Indications of a kick are generally an increase in pump speed, or an indication from the
Coriolis flow meter. Both separately and especially combined these indicators can early
indicate kicks of rather small sizes
Comment
This is done in the same way with or without the anti u-tube valve.
Step 3: Shut-in
(3-6)
The drillpipe is spaced out, choke line to BOP is opened, Valve 4 in Figure 9. Then the valve
to the CMP suction line is closed, Valve 1 in Figure 9. Now the well is prepared to be shut in,
and annular preventer is closed.
37 000,00
38 000,00
39 000,00
40 000,00
41 000,00
42 000,00
43 000,00
44 000,00
45 000,00
1,00 1,20 1,40 1,60 1,80 2,00 2,20 2,40
Pu
mp
ra
tes
[LP
M]
Mud weight [SG]
Pipe ID 3,75"
Cp=25
Cp=40
Figure 11: Freefall rate vs. mud weight, pipe ID 3.75”
38
Comment
The difference here is that a hard shut in is done, as the pumps aren’t shut down prior to
shut in. It says under step 4 in the document that: “The in-line choke valve (valve 2) and the
choke line to CMP pump valve (valve 3) are pre set when setting up for dual gradient
drilling”. What this means exactly is not fully understood, the procedures for a working anti
u-tube valve doesn’t express it in this way, merely telling which position the valves should be
in. It should say so in these procedures as well, so that the two different procedures don’t
have to be mixed up in a well control situation. By glancing at the next few steps, noting that
there isn’t said anything about these two valves, it is assumed that: Valve 2 (Figure 9) should
be closed to isolate rig choke line, and Valve 3 (Figure 9) should be opened to direct flow
from below the annular preventer and choke line to the CMP pump and MRL.
Step 4: Adjust pumps
(7-9)
The rig pumps should be slowed to kill rate, which is the fingerprinted rate that keeps 50 psi
(3.4 bar) on the standpipe. CMP pump suction pressure is increased to outflow matches
inflow, then 50 psi safety margin is added. The safety margin is well specific and may be
adjusted.
Comment
The pumps are synchronized to ensure the kill rate is uniform throughout the well. Whereas
the setup with anti u-tube valve demanded calculations to adjust the pump, this is based on
fingerprinting and adjusting the pumps to match. This can be easily done manually if the
automated system should fail. However, this is typically much like conventional well control,
and doesn’t offer any improvement to well control. The safety margin is the same as for
normal procedures and may be adjusted to well specific conditions such as safety towards
fracture pressure etc.
39
Step 5: Prepare to circulate
(10-13)
Kill calculations and decision on how to circulate the kill is done while the well is stable and
pumps are at kill rates. Use well control model to prepare kick circulation and do necessary
calculations. Decide based on the well control model if it is safe to bring the kick up to
surface through the CMP pump and MRL. Flow returns on the manifold is switched to the
degasser line rather than the Coriolis flow meter line. Automated choke control to keep
standpipe pressure and BHP constant. The CMP pump control is set to keep fixed speed,
thus keeping the kill rate regardless of well pressures. Now pressure control is by the
automated choke only
Comment
The procedure only says to use the well control model. Steps to do the calculations manually
are missing, and should be included for safety control and contingency should anything
about the automated system fail. Interesting to note that a comment about the decision to
circulate through CMP pump and MRL or not is made here. No such comment is made in the
normal procedures. In general the pressure control is better without the u-tube effect, but
the boiling of gas in the CMP pump and MRL is also a problem with the anti u-tube valve.
Awareness of this matter could with safety in mind be expressed in both procedures. Flow is
switched as with normal procedures. Automated choke control on the manifold the same as
for normal procedures. As for the normal procedures, CMP pump is operated at fixed speed,
giving control of the pressures to the automated choke. This result in a system generally
simple and similar to conventional drilling should manual control be necessary; adjustment
of the choke.
Step 6: Circulate out kick
(14-16)
The CMP suction pressure will indicate changes in the standpipe pressure and BHP, and will
allow for choke adjustment, given that CMP pump is operating at fixed speed. While
monitoring flow, continue till kick is out of the well
40
Comment
As with the normal procedures, the kick is circulated out with choke keeping the pressures
stable, and pumps keeping the kill rate. Missing again some thoughts and procedures to
handle the boiling of gas, by adjusting pressures and keeping the gas from boiling prior to
reaching the MRL.
Step 7: Confirm kick is out of the well
(16-17)
Confirm the well fingerprint to pre kick fingerprint
Comment
Should also include the comments from the normal procedure, where inspection of pits and
pit level as well as conferring with the whole crew to agree that kick is out of the well.
Otherwise the same as normal and conventional procedures.
Step 8: Flush BOP stack and circulate out trapped gas
(18-19)
There may be trapped gas below annular preventer, flush and circulate out
Comment
Standard procedure.
Step 9: Flow check and decide on method to balance pressure across BOP
(20-22)
The well is flow checked after CMP pump control has been switched back to suction
pressure control. Mud pumps are kept at pre kick kill rate, return flow should be on Coriolis
line. As with the normal procedures there are two methods to restore pressure across the
annular preventer.
Comment
In this step a flow check is performed to ensure a stable well, but the procedure doesn’t say
specifically to switch return flow on choke manifold, just to measure with Coriolis meter. Also
41
it should be noted that if the 50 psi wasn’t held at the standpipe, there might be air in the
drillstring, and this together with the possibility of u-tubing could result in a unsuccessful
flow check. Choosing method to restore pressure is the same as with normal procedures.
Method A
This method is similar to Driller’s method, as it uses heavier mud to keep the BHP constant,
while lowering the shut in casing pressure so that the pressure across the annular preventer
is equalized
Step A1: Set new mud weight and prepare new mud
(24-25)
The difference in pump suction pressure to ambient seawater pressure is the additional
pressure needed by the new mud weight. Calculate and prepare with safety margin.
Comment
Standard calculations and mud weight adjustment.
Step A2: Prepare CMP pump schedule
(26)
As the new heavier mud replaces old mud, suction pressure should be gradually decreased,
calculate and make schedule
Comment
Standard pump schedule, however u-tubing could affect the pressure here, so a safety
margin on the suction pressure could with benefit be added to compensate for this.
However, calculations shown above under freefalling may indicate that to eliminate u-tubing
by pumping may be hard, but the effect may be mitigated.
42
Step A3: Displace mud and check well
(27-29)
Displace mud, and when suction pressure equals or is close to ambient seawater (or blanket
fluid) pressure in the riser, the mud has been properly displaced. Flow check well by
stopping rig pumps and let well u-tube, check for flow from well.
Comment
New mud should provide desired BHP without the casing pressure, so suction pressure
should be close to ambient pressure. When the flow check is done, the mud is allowed to u-
tube. This will not increase BHP as the pump is set to operate on suction pressure. However,
this may be a very critical point if the flow check should fail. If there still should be gas in the
well, or a new influx occurs for any reason, the standpipe doesn’t have communication with
the mud in the drillstring. Thus the handling of a possible well control at this point may be
complicated, as the standpipe pressure can’t help determine the BHP.
Step A4: Additional check
(30-33)
Perform additional check to ensure that well is dead: stop CMP pump and monitor
casing\suction pressure, should there be a kick in the well, these will increase. If well is not
dead, recalculate and start procedure anew with new heavier mud.
Comment
Additional check to ensure well is dead. Since u-tubing already occurred with the previous
check it is safe to turn off CMP pump without risking an increase in BHP.
Resume operations if flow check ok.
Method B
As with the normal procedures, this method applies pressure on top of the annular
preventer to balance the pressure again.
Step B1: Calculate mud weight and prepare
(35-36)
43
Using the suction pressure again, calculate the required pressure supplied by mud and add
safety margin. With the new mud on top of the annular preventer, no casing pressure
should be needed when annular preventer is opened
Comment
Standard procedure calculating the mud weight from a given pressure requirement. The
procedure doesn’t say anything about balancing the mud weight against the volume (height)
of the mud column in the riser. As it will be hydrostatic pressure, a lower weight can be
chosen of the volume is greater, and vice versa.
Step B2: Use boost line to fill riser with mud
(37)
Fill the riser with mud to the required height to balance pressure and check pressures above
and below preventer.
Comment
The procedures states that mud should be pumped to the required height in the riser. This
should be corrected to clarify that the required volume should be pumped. The pressures are
checked with pressure transducers in the riser and in the CMP suction line
Step B3: Open annular preventer and set valves to correct position and flow check
(38-42)
Annular preventer and CMP suction line valve (Valve 1, Figure 9) is opened, while BOP choke
line valve (Valve 4, Figure 9) is closed. Now flow check well and check against fingerprinting
Comment
Here the u-tubing isn’t mentioned, but it should. Depending on the weight and volume of the
mud pumped into the riser in step B2, there could still be some u-tubing occurring during the
flow check. The procedure steps from Method A flow check might be put to good use here.
Resume operations if flow check is ok
44
6. Discussion
6.1. The CMP technology
As discussed in chapter 2.3 and 2.4 there are both advantages and challenges in using the
DGD technology.
The major advantage of the technology is the effect the removal of mud in the riser has on
the BHP, and the possibility to easily regulate it with the CMP subsea pump. The fluid
column is greatly changed by the fact that the pressure in the riser in general doesn’t affect
the BHP since it is equal to normal water pressure. The CMP pump can easily regulate the
BHP by adjusting the fluid column above the inlet to the suction line. Thus the drilling
window has been opened, since the gradients can be calculated from the mudline, and also
the response time for adjusting the BHP has been greatly improved.
This leads to making DGD and the CMP system a technology to enable the drilling of projects
with very tight drilling windows, and also improves drilling in normal operations. The
improvement is mainly from increasing the efficiency of drilling. With pressure control that
good, the BHP can be adjusted without weighting and circulating in new mud, so that lost
circulation and differential sticking should be very much less likely. This leads to less
nonproductive drilling time caused by these two incidents. Thus the system can prove to be
highly economical, and thus again another enabling aspect.
However, there are some obstacles to overcome for the technology to be widely accepted.
One of which may be that the system as it is designed, is highly dependent on the anti u-
tube valve. For the safe operation during shut-in or low pump rates, u-tubing will have to be
mitigated. The proposed anti u-tube valve is a means to meet this problem, but has yet to
be tested successfully in the field, according to Godhavn. He also proposes that a continuous
circulation device may help solve the problems related to u-tubing. U-tubing could, under
normal drilling operations affect the BHP, depending on what operation mode the CMP
pump is in. Should it be in constant suction pressure, it shouldn’t affect the BHP for too long
till the pump will adjust. Should it be in constant speed (constant flow) the BHP will increase
until u-tubing equalizes between the drillstring and annulus, or till it is stopped.
Furthermore, the u-tubing will result in air bubbles in the circulation system, which might
trigger the kick detection alarms, and they could also reduce the service time of the pump.
45
Therefore, either the u-tube valve has to be developed satisfactory, or other options to
prevent the u-tube effect should be considered.
The CMP system is a developed RMR system, as it is connected to the subsea BOP stack by a
riser, while the RMR is an open system. The added riser allows for drilling through
hydrocarbon bearing zones, as well control now is possible. The RMR system is designed for
tophole sections where well control is no issue. This results in that the mud return
equipment is not designed to handle well control, and it is to be anticipated that an attempt
to apply the same mud return equipment will be used for CMP purposes as well. Thus it is
not rated for the high pressures well control equipment normally is. For instance AGR
informs that the MRL is rated for operation on 35 bar. This is much less than a BOP is
required to handle. Testing and verification of the system’s ability to handle the pressures
associated with well control situations should be done. Furthermore it is worth to notice
that the MRL is a AGR design and is a rubber hose which features a ¼ turn connections. The
hose’s ability to potentially handle the gas coming out of solution can become a serious
issue.
Lastly, as with all new technologies, CMP will face a challenge in being implemented as a
new and different method to drill. CMP will not only be different in the way drilling is done,
it will also greatly affect the safety aspects of the drilling process. Although the system
should add to the safety of drilling for hydrocarbons, there is always great skepticism to
safety-changing technologies. This is both a good thing, and a demanding aspect. The rig
crews will have to develop new routines in well control, and train all over again on a new
system. As well control training is one of the prime training objectives of any drilling crew, it
will be even a greater challenge to train them in a new and different matter, and all the
while keeping confidence. This stresses the point of having very secure test results, just as
sound theoretical background, before the system is put to use in the field.
6.2. Automated well control
The information about the automated well control systems have been limited to whatever is
published material, and thus the information has also been very biased.
The two systems reviewed are both based on MPD drilling in their development. Although
the aim for the MFC system was stated to be for all drilling systems on any rig. Thus the real
46
technology here lies in the control system, which is the core of any automated operation.
Using a hydraulics model, together with interfaces for the available data, and taking default
inputs, the system will fully or partly automate some of the operations in a drilling system.
@Balance offers the DAPC system together with Schlumberger’s portfolio of drilling
services, while the MFC system is offered as a part of Weatherford’s drilling tools. That
makes them two competing systems offered by the two of the major service providers. Yet,
they are surprisingly similar, that they both offer the control system as the main service, and
rely heavily on the accurate flow measurements of the Coriolis meter. Although recent a
publication by VP of @Balance, Reitsma (2011) may indicate that soon, the means of
measuring flow may distinguish them from each other.
Other than that, it is hard to see what these systems contribute to drilling with. The
published material does argument that both systems have handled pressure control well,
and also handled (Roes et al. 2006; Fredericks et al. 2008; Helio Mauricio Santos et al. 2007).
Then again this is working together with other suppliers and MPD systems. As the CMP
system herein analyzed has its own control system and procedures for well control, it is hard
to see the applicability of these systems as standalone control-systems. An indication of this
might be that both systems are offered through service providers that also offers drilling
systems. However, the aim of the MFC system is an admiring one; to be applied to any
drilling system on any rig. That clear aim the @Balance’s DAPC system lacks. So the MFC
might add increased pressure control and add to the safety of conventional drilling. As the
advanced drilling system will be more applied, it would be beneficial to still drill
conventionally, but benefit from the increased safety of the more advanced systems,
without the full cost. This is a market that probably could prove successful as there are still
marginal fields that will have to be drilled with less costly conventional technologies, but still
satisfying increased demands for safety.
It remains to be seen if automated systems can be applied in this way, rather than being
bought by the big service providers. Because there is no doubt that any company that
provides MPD and DGD systems, will need a good controlling system. And with the broad
testing experience and database of such systems already tested, they may prove tempting
for such companies.
47
6.3. CMP well control
The steps for the CMP well control procedure have all been analyzed and reviewed in
Chapter 5. The discussion here will wrap up the general concerns of the procedure, and for
thought on each specific step, it is referred to the mentioned chapter.
CMP is an development of the RMR system which has been tested and prepared for field
tests. AGR currently have contracts to use the RMR system both in the Gulf of Mexico and in
Australia for deep water drilling. As earlier mentioned, what separates the two systems is
that the CMP system applies a riser and a BOP to allow for drilling in hydrocarbon bearing
geologies. Thus the systems feature to handle a well control situation hasn’t been tested, as
the CMP system is based on RMR, but not yet tested to full scale.
The RMR system is not designed for handling and circulating a kick, and the equipment
doesn’t have the pressure rating normal for well control handling equipment. As mentioned
the MRL hose is rated to a 35 bar pressure. This could prove problematic when there is a
chance of the influx of a kick to come out of solution in or before the riser. Generally, the
boiling point of gas in an oil based mud can be expected around 35 bar. This issue, and the
general issue of gas boiling is missing in the information provided by AGR. The issue of gas
coming out of solution will affect the pressures of the situation, and also the CMP pump will
become useless when filled with gas.
The procedure analyzed doesn’t either have any means of handling gas boiling. It seems like
the system was designed to handle a kick without addressing the matter. It will be able to
control the pressures, with good system control of pump rates and pressure applied by the
choke, but it will lack control of the point of boiling. This will generally prove problematic for
the operation of the pump, and possibly the integrity of the mentioned MRL hose.
Having both the CMP pump and the automated choke available as means of controlling
pressures in the circulation could prove for the control of also this problem. If the hose was
rated for handling the gas, the pump could modify the pressure in the annulus, without
bringing the BHP out of the drilling window, so that the boiling point is not reached before
the MRL hose. Then the choke could control the pressure loss in the MRL hose, and thus the
effect of boiling. If the pressures are right, it could also help hold the boiling point down
until the manifold and degasser is reached.
48
These are matters that could further strengthen the well control aspect of the CMP system.
However, simulations and tests should be done, to see if this is possible to achieve.
The procedure steps provided by AGR seem very well thought through, beside the fact that
they ignore the gas boiling problem.
The procedures are put for two system setups, with and without the anti u-tube valve. The
normal procedure being with the valve, and the contingency procedure for drillstrings
without the valve.
The normal procedure is in general similar to the driller’s method, in the way that the kick is
being circulated out of the well before the mud weight is adjusted. However, the BHP will at
once be adjusted to be over the kick pressure, and another kick shouldn’t happen. It is
circulated out with the rig pumps and CMP pump set to keep the kill rate uniform through
the well, and the automated choke to control the BHP as the kick progresses. When the kick
is out of the well, there are two options for the equalizing of the pressure across the BOP,
either to circulate in heavier mud, or add more mud on top of the BOP through a boost line
to the riser. The latter in general allows for any mud to be used, as long as the height of the
column and the mud weight applies the necessary pressure, while still ensuring it is
compatible with the original mud.
Should all the equipment work, the procedures are well functioning. They shouldn’t appear
as too different for the crew, as the general flow of actions is the same as with conventional
well control. However, an important thing to point out is that it is suggested that a soft shut
in procedure is applied, though that shouldn’t be unknown to the crews, despite not being
the standard procedure. Also they do explain the calculations needed, and should help the
crew to keep up with and follow the automated system. This requires the control system to
also display the data used for the automated control, so that manual follow-up is possible.
Should the system fail however, it may become difficult to manually control the system in a
way that yields the same control over BHP and flow as the automated system would. This
might be a reason for keeping the CMP pump on constant speed\flow, so that the focus in
the event of system failure would be to control the choke, as with conventional well control.
49
Then again, a specific procedure for how to control the well should the control system
somehow malfunction is not provided. The procedures clearly state that should a problem
occur, the rig pumps should be stopped to avoid fracturing the well, and then ensure the
BHA is pulled up into the last casing, so equipment can safely be repaired. Since CMP is a
new system without field tests, it is reasonable to believe that possible problems will have
to be met and solved, for the gaining of experience as a base for procedures. However, two
contingency plans should be made beforehand without the need for field experiences: A
procedure on how to control the well conventionally through the rig choke line, in the event
of failure on CMP pump, MRL or the CMP choke manifold, and a procedure on how to
control the well manually, in the event of failure on the CMP control system. The control of
the well could be complex, with pump, choke, and possibly point of gas boiling to control.
This requires strict and clear procedures to maintain the safety of personnel and equipment.
This could be a point where CMP well control could prove to be more complex than
conventional well control, and if gas is in the MRL, conventional well control is no longer an
option. Thus the operation might be less safe and controlled than conventional control. That
would be a major setback and obstacle for the implementation for CMP drilling. It is not
likely that technologies that in the worst case could cause a lessened safety would be
broadly applied, or even permitted in most areas.
50
7. Conclusions
From the analysis it can be concluded that there are both benefits and challenges in
implementing CMP, both regarding the method itself, but most importantly the well control
of the system.
• The procedures lack some key points
o What to do in the event of control system failure?
o How to manually calculate and monitor control system?
• Both the procedures and the available material fail to address the important
problem of gas boiling
• The method for safe well control will heavily rely on the u-tube valve. Without it, the
well control will still be possible, but not safer than conventional well control
• The mud return equipment, hose and pump, need to be proven to be able to handle
the possible pressures of a well control situation, as well as gas.
• The CMP system has the potential to become both an enabling and improving
technology. If the well control issues pointed to herein, it will also increase the safety
of drilling.
• In the implementation the CMP system, well control will be the major obstacle
• Training personnel and gaining their trust and confidence with the system have to be
prioritized.
• @Balance’s DAPC control system is a promising candidate for increase well control,
but should be coupled with a MPD system for successful implementation
• Secure Drilling’s MFC system has an ambition and prospect of becoming an industry
standard that improves safety without the need of great modifications
• DAPC could have the same opportunity, but lacks the ambition and goal which could
prove MFC successful.
51
7.1. Further work
This work has been done purely analytical, and further work should be carried out to
evaluate the well control procedures for CMP, as well as the potential for the automated
well control systems MFC and DAPC
• Simulate and model the well control procedures of CMP
• Specially look into the issue with gas boiling, do simulations to prove if it will be a
problem or not, and decide how the system should handle it.
• Lab tests of the well control abilities of the CMP should follow the simulations, to
look for possible caveats in the procedures.
• Field tests of the CMP technology, and the well control abilities
• Follow up from the RMR operations to look for improvements that can be made to
the equipment
• Consider options to the anti u-tube valve if it cannot be made to work correctly
52
8. Nomenclature
9.
BHA: bottomhole assembly;26
BHP: bottomhole pressure;7
BOP: Blowout preventer;III
BU: bottoms up;13
CCS: Continuous circulation coupler;17
CMP: Controlled mud pressure;III
DAPC: Dynamic annular pressure control;III
DGD: Dual gradient drilling;III
FCP: Final circulating pressure;16
ICP: Iinitial circulating pressure;16
LPM: litres per minute;36
MFC: Microflux control;III
MPD: Managed pressure drilling;III
MRL: Mud return line;III
NCS: Norwegian continental shelf;25
�_7=>?, �@> � ABCA8: Original set suction pressure;27
�_ABCA: Kick pressure;27
�_D�>E: Opening pressure of anti u-tube valve;27
�_=FG>?H: Safety margin;27
�_=>?: Set suction pressure;27
Pc1: Ffriction loss, original mud;16
Pc2: Friction loss, kill mud;16
PWD: Pressure while drilling;17
RCD: Rotating circulation device;17
SCR: Pressure loss at slow circulation rate;16
SG1: Original mud weight;16
SGkill: kill mud weight;16
SICP: Shut in casing pressure;15
SIDPP: Shut in drillpipe pressure;14
SPM: subsea pump module;4
t1: Surface to bit travel time;16
t2: Bit to surface travel time;16
TVD: True vertical depth;15
53
10. References
Personal communication with John-Morten Godhavn, Nils Lennart Rolland and Kjell-Rune
Toftevåg.
AGR procedure documentation, in procedure document and slideshow.
Bourgoyne, Adam T. 1986. Applied drilling engineering. Vol. 2. SPE textbook series.
Richardson, TX: Society of Petroleum Engineers.
Forrest, Neil, Tom Bailey, and Don Hannegan. 2001. Subsea Equipment for Deep Water
Drilling Using Dual Gradient Mud System. In SPE/IADC Drilling Conference. Amsterdam,
Netherlands.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=00067707&societyCode=S
PE.
Fredericks, Paul Douglas, Don Reitsma, Tom Runggai, James Neil Hudson, Ralf Zaeper, Oliver
Backhaus, and Maximo Hernandez. 2008. Successful Implementation of First Closed Loop,
Multiservice Control System for Automated Pressure Management in a Shallow Gas Well
Offshore Myanmar. In IADC/SPE Drilling Conference. Orlando, Florida, USA.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-112651-
MS&societyCode=SPE.
Frøyen, Johnny, Nils Lennart Rolland, Rolv Rommetveit, Roger Stave, and Hitesh Jaising.
2006. Riserless Mud Recovery (RMR) System evaluation for top hole drilling with Shallow
Gas. In SPE Russian Oil and Gas Technical Conference and Exhibition. Moscow, Russia:
Society of Petroleum Engineers.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-102579-
MS&societyCode=SPE.
Gonzalez, Romulo. 2001. Deepwater drill string shut-off valve system and method for
controlling mud circulation. July 24. http://patft.uspto.gov/netacgi/nph-
Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.ht
m&r=1&f=G&l=50&s1=6263981.PN.&OS=PN/6263981&RS=PN/6263981.
54
Gonzalez, Romulo, and Frans Sippo Wiegand Smits. 2002. Deepwater drill string shut-off.
June 11. http://patft.uspto.gov/netacgi/nph-
Parser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.ht
m&r=1&f=G&l=50&s1=6401823.PN.&OS=PN/6401823&RS=PN/6401823.
Grace, Robert D., Bob Cudd, Richard S. Carden, and Jerald L. Shursen. 2003. Blowout and
Well Control Handbook. Gulf Professional Publishing.
Oskarsen, R.T. 2001. Toolkit and Drillstring Valve for Subsea Mudlift Drilling. College Station,
Texas: Texas A&M U.
Rehm, Bill, Jerome Schubert, Arash Haghshenas, Amir Saman Paknejad, and Jim Hughes.
2008. Managed Pressure Drilling. Gulf Publishing Company.
Reitsma, Don. 2010. A simplified and highly effective method to identify influx and losses
during Managed Pressure Drilling without the use of a Coriolis flow meter. In SPE/IADC
Managed Pressure Drilling and Underbalanced Operations Conference and Exhibition. Kuala
Lumpur, Malaysia: SPE/IADC Managed Pressure Drilling and Underbalanced Operations
Conference and Exhibition.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-130312-
MS&societyCode=SPE.
Reitsma, Don. 2011. Development of an Automated System for the Rapid Detection of
Drilling Anomalies using Standpipe and Discharge Pressure. In SPE/IADC Drilling Conference
and Exhibition. Amsterdam, The Netherlands.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-140255-
MS&societyCode=SPE.
Roes, Vincent C., Don Reitsma, Larry Smith, John McCaskill, and Fred Hefren. 2006. First
Deepwater Application of Dynamic Annular Pressure Control Succeeds. In IADC/SPE Drilling
Conference. Miami, Florida, USA.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-98077-
MS&societyCode=SPE.
55
Santos, H., P. Reid, and A. Lage. 2004. Opening New Exploration Frontiers With The Micro-
Flux Control Method For Well Design. In .
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=OTC-16622-
MS&societyCode=OTC.
Santos, Helio Mauricio, Erdem Catak, Joseph Irvin Kinder, and Paul Sonnemann. 2007. Kick
Detection and Control in Oil-based Mud: Real Well Test Results Using Micro-Flux Control
Equipment. In SPE/IADC Drilling Conference. Amsterdam, The Netherlands.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=SPE-105454-
MS&societyCode=SPE.
Santos, Helio, Christian Leuchtenberg, and Sara Shayegi. 2003a. Micro-Flux Control: The
Next Generation in Drilling Process for Ultra-deepwater. In .
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=OTC-15062-
MS&societyCode=OTC.
Santos, Helio, Christian Leuchtenberg, and Sara Shayegi. 2003b. Micro-Flux Control: The
Next Generation in Drilling Process. In SPE Latin American and Caribbean Petroleum
Engineering Conference. Port-of-Spain, Trinidad and Tobago.
http://www.onepetro.org/mslib/app/Preview.do?paperNumber=00081183&societyCode=S
PE.
White, Frank M. 2008. Fluid mechanics. Boston: McGraw-Hill.
56
11. Appendix
Appendix A: Worksheet Wait and weight, surface BOP: A-C
Appendix B: Worksheet Wait and weight, subsea BOP D-F
Appendix C: Driller’s method G-I
A
Appendix A: Worksheet Wait and weight, surface BOP
B
C
D
Appendix B: Worksheet Wait and weight, subsea BOP
E
F
G
Appendix C: Driller’s method
H
I